Model comparison
Gemini 1.5 Pro (May 2024) vs GPT-4
Gemini 1.5 Pro (May 2024) is the stronger model overall, scoring 32.1 to 29.1 on the Noometry Index.
Last verified . 29 shared benchmarks.
GoogleSummary
- They share 29 benchmarks with published results for both. Gemini 1.5 Pro (May 2024) scores higher in 7 categories and GPT-4 in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Gemini 1.5 Pro (May 2024) leads 52.4 to 34.9.
- The biggest single-benchmark swing is MATH Level 5: 70.4% for Gemini 1.5 Pro (May 2024) and 23% for GPT-4.
Side by side
| Gemini 1.5 Pro (May 2024) | GPT-4 | |
|---|---|---|
| Provider | OpenAI | |
| Noometry Index | 32.1 | 29.1 |
| Released | 2024-02-15 | 2023-03-14 |
| Weights | Proprietary | Proprietary |
| Context window | — | 8K |
| Max output | — | 8K |
| Input $ / M tokens | — | $30 |
| Output $ / M tokens | — | $60 |
| Results tracked | 45 | 38 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 34.2 (#241), GPT-4: 31.6 (#283)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| WeirdML | 22.2% | 12.4% |
| BigCodeBench Instruct | 43.8% | 46% |
| LMArena Coding | 1294 | 1254 |
| BigCodeBench Complete | 57.5% | 57.2% |
| HumanEval+ | 79.3% | 79.3% |
| CadEval | 34% | — |
| MBPP+ | 74.6% | — |
Agentic & Tool Use Not comparable
Gemini 1.5 Pro (May 2024): 17.9 (#145), GPT-4: —
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| TheAgentCompany | 3.4% | — |
| Cybench | 7.5% | — |
| BALROG | 21% | — |
| METR Time Horizons | — | 36.1% |
Reasoning GPT-4 leads
Gemini 1.5 Pro (May 2024): 12.3 (#338), GPT-4: 17.8 (#289)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| LMArena Hard Prompts | 1296 | 1241 |
| DTBench | 59% | 62.7% |
| BIG-Bench Hard | 89.2% | 75.1% |
| Epoch Capabilities Index | 131.73 | 125.89 |
| ForecastBench | 58.4 | 57.8 |
| ARC-AGI-2 | 0.8% | — |
| SimpleBench | 27.1% | — |
| Chess Puzzles | — | 4% |
| Mystery Game Puzzles | — | 12% |
| LMCA | — | 17.1% |
| HellaSwag | — | 95.3% |
| WinoGrande | — | 87.5% |
Math Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 25.8 (#266), GPT-4: 10.8 (#309)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 23.1% | 1.1% |
| LMArena Math | 1315 | 1269 |
| MATH Level 5 | 70.4% | 23% |
| Omni-MATH | 36.4% | — |
| GSM8K | — | 92% |
Knowledge Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 29.4 (#239), GPT-4: 18.4 (#282)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| GPQA Diamond | 57.2% | 35.7% |
| LMArena Expert | 1279 | 1211 |
| MMLU | 86.9% | 86.4% |
| Humanity's Last Exam | 4.6% | — |
| MMLU-Pro | 73.7% | — |
| Confabulations | 13.5% | — |
| GPQA (HELM) | 53.4% | — |
| TriviaQA | — | 84.8% |
Multimodal Not comparable
Gemini 1.5 Pro (May 2024): 36.8 (#77), GPT-4: —
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| LMArena Vision | 1161 | — |
| Video-MME | 75% | — |
Multilingual Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 45.3 (#174), GPT-4: 40.6 (#215)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| LMArena Non-English | 1312 | 1246 |
| LMArena Chinese | 1331 | 1242 |
| LMArena French | 1302 | 1283 |
| LMArena German | 1286 | 1251 |
| LMArena Japanese | 1292 | 1209 |
| LMArena Korean | 1298 | 1184 |
| LMArena Russian | 1320 | 1251 |
| LMArena Spanish | 1311 | 1261 |
Instruction Following Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 68.6 (#185), GPT-4: 65.3 (#222)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| LMArena Instruction Following | 1297 | 1241 |
| IFEval | 83.7% | — |
Long Context Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 39.8 (#169), GPT-4: 37.7 (#212)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| LMArena Longer Query | 1308 | 1244 |
Writing & Preference Gemini 1.5 Pro (May 2024) leads
Gemini 1.5 Pro (May 2024): 52.4 (#172), GPT-4: 34.9 (#268)
| Benchmark | Gemini 1.5 Pro (May 2024) | GPT-4 |
|---|---|---|
| LMArena Text | 1319 | 1263 |
| LMArena Creative Writing | 1333 | 1244 |
| LMArena Multi-Turn | 1296 | 1257 |
| EQ-Bench Creative Writing | — | 752 |
| WildBench | 81.3% | — |
Frequently asked questions
Is Gemini 1.5 Pro (May 2024) better than GPT-4?
Gemini 1.5 Pro (May 2024) is the stronger model overall, scoring 32.1 to 29.1 on the Noometry Index.
Is Gemini 1.5 Pro (May 2024) or GPT-4 better for coding?
Gemini 1.5 Pro (May 2024) scores higher on coding benchmarks: 34.2 versus 31.6 in the Noometry coding category.
How many benchmarks do Gemini 1.5 Pro (May 2024) and GPT-4 share?
29 benchmarks have published results for both models. Gemini 1.5 Pro (May 2024) has 45 scored results on Noometry and GPT-4 has 38.